Construction and Case Analysis of a Cocooning Degree Measurement Model for Online New Media Information

In the era of online new media, algorithmic recommendations are widely employed, giving rise to the phenomenon of information silos. This has adverse effects on user cognition and the online ecosystem, making quantitative research on this topic of significant practical importance. This study collected user behaviour and content data from multiple self-media platforms, Kuaishou, and news platforms. It analysed the characteristics of self-media content, the mechanisms behind the formation of information silos, and the reinforcing role of algorithms. It also compared the silo behaviour exhibited by short video and news platforms. A measurement model for information silos was constructed based on Shannon's information entropy. Using collected user browsing data and recommendation information from news platforms, empirical calculations were performed to analyse overall user silo characteristics, differences in silo effects among users of varying activity levels, and the evolution of silo trends across different phases of the Kuaishou platform. Findings reveal that while information echo chambers exist across new media platforms, they do not form closed barriers. Higher information entropy values correlate with weaker echo chamber effects, and user activity levels exhibit a non-linear relationship with echo chamber intensity. Algorithm optimisation significantly mitigates echo chamber effects. This study's measurement model provides a basis for assessing echo chamber intensity, while its conclusions offer strategic support for mitigating information echo chambers.